Objectives. Although menopausal hormone replacement therapy (HRT) is widely recommended for treatment of menopausal symptoms and prevention of osteoporosis, its effects on the risk of breast cancer are not clear. Therefore we propose to conduct a population-based case-control study focusing on the associations of various HRT preparations with breast cancer. Sweden offers unique opportunities to study HRT and breast cancer because of its nationwide cancer registration, population registers, high participation rate in epidemiologic studies and widespread use of estrogen (estradiol and conjugated estrogens) and estrogen-progestin treatment. Specific aims. Specific hypotheses to be addressed include: 1) Treatment with potent estrogens causes an increased risk of breast cancer which is related to duration of use and latency since first use and which persists even when treatment has been discontinued. 2) Estradiol has a stronger adverse effect on breast cancer risks than conjugated estrogens. 3) Use of estrogen plus progestin increases the risk of breast cancer more than estrogen alone. Experimental design and methods. A nationwide population-based case- control study is proposed, building on the experience of the investigators in similar large-scale studies. Eligible as cases will be all Swedish females aged 50-74 years diagnosed with histopathologically confirmed invasive breast cancer during the study period. Over 18 months, approximately 4,250 incident cases will be diagnosed and notified to the Cancer Epidemiology Unit. Controls will be selected - stratified by 5-year age group - from a national population register to form a control group of similar size as the case group. A covering letter, a questionnaire and a booklet with color photographs of all HRT brands used in Sweden will be mailed to all potential cases and controls. Reminders by mail and complementary telephone interviews are anticipated to provide at least an 80% participation rate among controls and 80% among cases, i.e. 3,400 cases and 3,400 controls. At the age of 60 years or more, 18- 20% of the controls are estimated to be ever users of HRT and 50% of these women will have used combined estrogen progestin treatment. This exposure frequency in the study population would yield an 80% power (alpha = 0.05, two-tailed) to detect a 20% - and a 99% power to detect a 50% - or more increased risk of breast cancer among women ever exposed to estradiol or conjugated estrogens and similar power among ever users of estrogen-progestin in combination. The corresponding estimates by duration of combined treatment were 84% for each stratum of 1-6, 6-11 and 12+ years respectively.